Memory, without the privacy tax

AI that remembers you is convenient. It's also, increasingly, a liability.

Over the past year, persistent chatbot memory has gone from a headline feature to a documented risk. The Wall Street Journal has flagged the privacy exposure of always-on memory. A 2026 CHI study found users unsettled when ChatGPT recalled personal details "with such accuracy," describing it as "a little bit intrusive" and asking for more control over what gets kept. Security researchers have named a new attack class—memory poisoning—where malicious content written into an agent's long-term memory keeps steering its answers in every future session. And in June 2025, a court order required OpenAI to retain certain ChatGPT content, meaning even opt-out settings could be overridden by an obligation users never saw.

The lesson isn't "memory is bad." It's that memory should be built privacy-first, on infrastructure that can't quietly read or leak what it stores.

That's how we built PayPerQ Memory.

What it is

PayPerQ Memory lets the assistant carry useful context across conversations—your preferences, your projects, the facts you'd otherwise repeat every time—so you don't have to reintroduce yourself with each new chat.

What makes it different is where that memory lives and how it's produced:

• Local-first by default. Your memory is stored in your own browser. Nothing is synced to a server unless you explicitly turn sync on.
• Extracted inside a confidential enclave. When PayPerQ decides what's worth remembering, that step runs through the same end-to-end encrypted confidential-computing pipeline we use for private inference. Your conversation is end-to-end encrypted from your device to the enclave, processed inside a hardware-backed Trusted Execution Environment (AMD SEV-SNP / Intel TDX with confidential GPUs), and the PayPerQ proxy in the middle stays blind to the contents.
• Encrypted when you sync. Turn on a passkey and your synced memory is end-to-end encrypted—stored as ciphertext that only your key can unlock. We hold the blob; we can't read it.
• Fully under your control. You decide what's remembered and can delete it at any time.

Key capabilities

• Cross-chat recall — the assistant remembers stable facts and preferences across separate conversations.
• Per-conversation toggle — flip memory On or Off for any single chat with one tap, so a sensitive conversation never touches your memory.
• Confidential extraction — the "what should I remember?" step happens inside a TEE; the raw conversation is never exposed to the host, the cloud provider, or PayPerQ.
• Optional encrypted sync — off by default; when enabled with a passkey, memory is end-to-end encrypted before it leaves your device.
• Granular management — view everything that's stored, delete individual items, multi-select and bulk-delete, or clear it all. Turning memory off can wipe every saved fact from both your device and the server in one action.

A few ways it helps in practice:

• Tell it once that you write in British English and prefer concise answers—every future chat honors it.
• Keep a running project context ("I'm building a Next.js app on Supabase") without re-pasting the stack each session.
• Run a private, memory-off chat for anything you don't want retained, while your normal chats stay personalized.
• Clear your entire memory before handing your laptop to someone else—one click, gone from device and server.

How to Use on PayPerQ

What is PayPerQ?

PayPerQ is a pay-per-query AI service that gives you instant access to hundreds of chat, image, video, and audio AI models in one place. Unlike traditional ChatGPT subscription that charges $20+ per month, PPQ users pay only for what they use—averaging just $4 a month.

Account registration optional. No monthly commitments. Privacy focused. Credit cards and all major cryptos accepted. Just top up with as little as 10 cents and start using premium AI immediately.

Why Use PayPerQ?

• Access hundreds of AI models from all major providers in one place
• Pay per use - no subscriptions, no wasted money on unused credits
• No registration required - start using AI in seconds
• Start small - top up with as little as 10 cents
• Privacy-first - conversational data stored locally by default
• Average cost: ~1 cent per query

Getting Started

1. Visit ppq.ai
2. Top up your balance (crypto or credit card, as little as 10 cents)
3. Select your model and start chatting

No account creation needed—just fund and go.